Hans-Joachim Lutz’s research while affiliated with European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) and other places

The objective of this study is to improve the characterization of satellite-derived atmospheric motion vectors (AMVs) and their errors to guide developments in the use of AMVs in numerical weather prediction. AMVs tend to exhibit considerable systematic and random errors that arise in the derivation or the interpretation of AMVs as single-level point observations of wind. One difficulty in the study of AMV errors is the scarcity of collocated observations of clouds and wind. This study uses instead a simulation framework: geostationary imagery for Meteosat-8 is generated from a high-resolution simulation with the Weather Research and Forecasting regional model, and AMVs are derived from sequences of these images. The forecast model provides the truth with a sophisticated description of the atmosphere. The study considers infrared and water vapor AMVs from cloudy scenes. This is the first part of a two-part paper, and it introduces the framework and provides a first evaluation in terms of the brightness temperatures of the simulated images and the derived AMVs. The simulated AMVs show a considerable global bias in the height assignment (60-75 hPa) that is not observed in real AMVs. After removal of this bias, however, the statistics comparing the simulated AMVs with the true model wind show characteristics that are similar to statistics comparing real AMVs with short-range forecasts (speed bias and root-mean-square vector difference typically agree to within 1 m s(-1)). This result suggests that the error in the simulated AMVs is comparable to or larger than that in real AMVs. There is evidence for significant spatial, temporal, and vertical error correlations, with the scales for the spatial error correlations being consistent with estimates for real data.

The work described here is part of a wider study whose main objective is to improve the characterization of Atmospheric Motion Vectors (AMVs) and their errors to improve the use of AMVs in Numerical Weather Prediction (NWP). AMVs are estimates of atmospheric wind derived by tracking apparent motion across sequences of meteorological satellite images, and it is known that they tend to exhibit considerable systematic and random errors and geographically varying quality, as shown in comparisons against radiosonde or NWP data. However, there is a rather limited knowledge of the characteristics and origin of these errors, although the height assignment is generally recognized as a key source of error. An important difficulty in the study of AMV errors is the scarcity of collocated observations of cloud and wind. To overcome that difficulty, we approach the analysis of AMV errors using a simulation framework: geostationary imagery is generated from a high resolution NWP model simulation, and AMVs are derived from sequences of simulated images. The NWP model provides a " ground truth " , which allows a detailed study of AMV errors, bypassing the usual difficulty of the scarcity of observations. Provided model simulations are realistic, the analysis of AMV errors in this setting can shed light on the nature of AMVs derived from observed imagery and their errors. The study is performed on the basis of Meteosat-8 simulations from the Weather Research and Forecasting (WRF) regional model, which has a nominal horizontal resolution of 3 km. This paper focusses on the part of the study that explores the assignment of AMVs to levels related to model clouds. Section 1 introduces the paper, and section 2 gives a brief description of the NWP model simulation and the AMV derivation. Section 3 presents comparisons between the simulated AMVs and the model " true " winds, exploring several possible ways of attributing height to the AMVs. Section 4 focusses on multi-layer situations, highlighting some of the problems associated, and illustrating how a simulation framework may help to study these situations. Finally, section 5 presents the conclusions.

The main objective of this study is to improve the characterization of Atmospheric Motion Vectors (AMVs) and their errors to improve the use of AMVs in Numerical Weather Prediction (NWP). It is known that AMVs tend to exhibit considerable systematic and random errors and geographically varying quality, as shown in comparisons against radiosonde or NWP data. However, there is a rather limited knowledge of the characteristics and origin of these errors: they can arise in the AMV derivation process, but they can also arise from the interpretation of AMVs as single-level point observations of wind. An important difficulty in the study of AMV errors is the scarcity of collocated observations of clouds and wind. To overcome that difficulty, this study approaches the analysis of AMV errors using a simulation framework in which AMVs are derived from sequences of images simulated from atmospheric forecast model data. In this framework the model provides a "ground truth", including wind and cloud distributions, which allows a detailed study of AMV errors. Provided model simulations are realistic, the analysis of AMV errors in this setting can shed light on the nature of AMVs derived from observed imagery and their errors. The model used for the simulation is the Weather Research and Forecasting (WRF) regional model, and the nominal horizontal resolution of the simulation is 3km. This presentation shows the main results of the ongoing study. First, cloud structures from observed and simulated images are compared. Then AMVs, interpreted as single-level point estimates of wind, are evaluated by comparison to the model truth. Then we present results regarding horizontal, vertical and temporal error correlations. Finally, we evaluate AMVs interpreted as vertical and horizontal averages of wind.

The knowledge of the surface emissivity is an important condition for all IR radiative transfer (RT) calculations and thus important for all satellite products that are based on RT calculations. A variety of the Meteosat Second Generation (MSG) algorithms use techniques which are based on the RT results, e.g. the cloud and dust storm detection , the active fire monitoring, the derivation of instability indices, and of total columnar ozone content. Emissivity information has been derived from the University of Wisconsin - CIMSS global infrared land surface emissivity database with its high spectral and high spatial resolution. This data has been re-mapped in space and spectra to the Meteosat-8/9 SEVIRI channel definition. This poster presents the inclusion of the emissivity information in the product generation process, and its impact on the final products. 1. BASIC PRINCIPLE

We developed a method to retrieve the optical thickness (tau) and effective particle radius (r(e)) of water clouds using the split-window channels and the 8.7-mu m channel of Meteosat-8. Valid ranges are approximately from 1 to 9 for tau and smaller than 18 mu m for r(e). Water clouds were first identified using 8.7-mu m and 11-mu m data. The retrieval used the brightness temperature (TBB) and brightness temperature difference (BTD) between the split windows, as computed with the radiation code RSTAR5b for various properties of water clouds and vertical profiles of temperature and water vapor. Arch-shaped curves in the TBB and BTD domains resembled those for ice clouds, depending on tau and r(e). The retrieved cloud parameters were then compared to those retrieved by the solar reflection method, which uses the 0.6-, 3.9-, and 11-mu m channels of Meteosat-8. Comparison of the two methods revealed that the split-window technique could capture spatial features for both tau and r(e), showing some agreement with r(e) for the solar reflection but needing more improvement for tau in the current case study.

there was no attempt to establish an absolute reference. Here we report on selected results. A more complete description of the results will be made available through our webpages. 2 Observations For the present work MVIRI-7 observations were analysed. Being an instrument mounted on a geostationary platform, this means that a larger area can be observed within a relatively short period of time (30 minutes). It also means that the spatial resolution of the observations is limited (4.8 km at sub-satellite point). MVIRI7 is a three channel radiometer, which means that under certain situations the detection of clouds is quite uncertain. Four periods of one week equally distributed throughout the year were selected to document any temporal dependency of the findings. Figuur 1 Time series of cloudfree fraction for Europe for four weeks equally distributed throughout 1999. Namely the first week in Jan 1999 (upper left panel), first week in April 1999 (upper right panel), first week of

Results of the split-window cloud retrieval method and the new Meteosat Second Generation cloud analysis method (MSG/CLA), have been compared for MODIS data over the west Atlantic Ocean. Very good agreement is obtained for the classification of optically thick ice and water clouds. Differences are found for thin cirrus, thin water clouds and at cloud edges. These differences are explained by the fact that MSG/CLA also uses spectral channels of 3.9, 6.2, and 8.7 m in addition to the split-window, which provides information over and above the split-window observations. Some of the disagreement at cloud edges is interpreted as inter-channel miss-alignment. The analysis in this study also confirms that an optically thin water cloud can be correctly classified by the MSG/CLA method.

For over two decades operational rainfall estimations from geostationary satellites have represented an ambitious aspiration of scientists and an identified need of operational meteorologists. A wide variety of infrared and combined visible and infrared methods have been proposed for the identification of suitable relationships between satellite-observed cloud top radiative features and rainfall at the ground. Microwave-based retrievals, however, correlate rainfall and internal cloud microphysical features more successfully. The most significant limitation, however, is the indirect character of the retrieval that correlates microphysical and dynamical cloud characteristics with rain amounts at ground level. METEOSAT Second Generation signals a new era for geostationary satellites with its new 12 channel imager SEVIRI and 15 minute full-disk image repeat cycle. SEVIRI is expected to contribute significantly to a better characterisation of clouds and atmospheric stability by means of improved infrared calibration, radiometric performances, imaging frequency and multispectral image analysis. The significant increase of multispectral cloud observations is expected to provide new data for the improvement of rainfall estimations from geostationary orbit. The anticipated progress from enhanced imaging frequency and multispectral data for the definition of new techniques is discussed. Considerations for operational applications, chiefly for nowcasting, are also provided as they are the main goal of the satellite. Future developments and synergies with other geostationary and polar orbiting instruments, passive and active, are finally considered as the ultimate strategy for more accurate instantaneous rainfall estimations at all latitudes. Copyright © 2001 Royal Meteorological Society

The objective of this study is to identify potential problems of and increase confidence in the operational calibration of the water vapor channel of the Meteosat Visible and Infrared Radiation Imager (MVIRI). First, the operational vicarious calibration method of the MVIRI water vapor channel is analyzed. This is done through a comparison of the operational calibration coefficient with a calibration coefficient derived from collocated satellite observations taken by water-vapor-sensitive channel 12 of the High Resolution Infared Sounder (HIRS) on NOAA 14. The analysis suggests a relatively large difference between the two calibration coefficients, the operational calibration coefficient being 10-15% larger than the one from the intersatellite calibration. Second, no such difference is found to exist between the MVIRI operational calibration coefficient and an intersatellite calibration coefficient derived from collocated GOES 8 water-vapor-sensitive channel 3 observations. Third, it is shown that the operational calibration coefficient for the MVIRI water vapor channel on Meteosat 7 is consistent with its blackbody calibration. At the present stage of the research no definite conclusions can be made. The results support the previously documented accuracy of 5% for the operational calibration coefficient of the MVIRI water vapor channel.